1. Field of the Invention
The present invention relates to a plasma processing apparatus, a plasma processing system, a performance validation system, and an inspection method therefor. The present invention can be suitably applied to a plasma processing apparatus having a plurality of plasma processing units so as to minimize the variation among the plurality of the plasma processing chambers and to improve the deposition characteristics using a power of higher frequencies.
2. Description of the Related Art
FIG. 33 illustrates a typical conventional dual-frequency excitation plasma processing unit which constitutes a plasma processing apparatus and performs a plasma process such as a chemical vapor deposition (CVD) process, a sputtering process, a dry etching process, or an ashing process.
In the plasma processing unit shown in FIG. 33, a matching circuit 2A is connected between a radiofrequency generator 1 and a plasma excitation electrode 4. The matching circuit 2A matches the impedances of the radiofrequency generator 1 and the excitation electrode 4.
Radiofrequency power generated from the radiofrequency generator 1 is supplied to the plasma excitation electrode 4 through the matching circuit 2A and a feed plate 3. The matching circuit 2A is accommodated in a matching box 2 which is a housing composed of a conductive material. The plasma excitation electrode 4 and the feed plate 3 are covered by a chassis 21 made of a conductor.
An annular projection 4a is provided on the bottom face of the plasma excitation electrode (cathode) 4, and a shower plate 5 having many holes 7 comes into contact with the projection 4a below the plasma excitation electrode 4. The plasma excitation electrode 4 and the shower plate 5 define a space 6. A gas feeding tube 17 comprising a conductor is connected to the space 6 and is provided with an insulator 17a at the middle thereof so as to insulate the plasma excitation electrode 4 and the gas source.
Gas from the gas feeding tube 17 is introduced inside a plasma processing chamber 60 composed of a chamber wall 10, via the holes 7 in the shower plate 5. An insulator 9 is disposed between the chamber wall 10 and the plasma excitation electrode (cathode) 4 to provide insulation therebetween. The exhaust system is omitted from the drawing.
A wafer susceptor (susceptor electrode) 8 which holds a substrate 16 and also functions as another plasma excitation electrode is installed inside the plasma processing chamber 60. A susceptor shield 12 is disposed under the wafer susceptor 8.
The susceptor shield 12 comprises a shield supporting plate 12A for supporting the susceptor electrode 8 and a supporting cylinder 12B extending downward from the center of the shield supporting plate 12A. The supporting cylinder 12B extends through a chamber bottom 10A, and the lower portion of the supporting cylinder 12B and the chamber bottom 10A are hermetically sealed with bellows 11.
The shaft 13 and the susceptor electrode 8 are electrically isolated from the susceptor shield 12 by a gap between the susceptor shield 12 and the susceptor electrode 8 and by insulators 12C provided around the shaft 13. The insulators 12C also serve to maintain high vacuum in the plasma processing chamber 60. The susceptor electrode 8 and the susceptor shield 12 can be moved vertically by the bellows 11 so as to control the distance between plasma excitation electrodes 4 and the susceptor electrode 8.
The susceptor electrode 8 is connected to a second radiofrequency generator 15 via the shaft 13 and a matching circuit accommodated in a matching box 14. The chamber wall 10 and the susceptor shield 12 have the same DC potential.
FIG. 34 illustrates another conventional plasma processing unit. Unlike the plasma processing unit shown in FIG. 33, the plasma processing unit shown in FIG. 34 is of a single-frequency excitation type. In other words, radiofrequency power is supplied only to the cathode electrode 4 and the susceptor electrode 8 is grounded. Moreover, the matching box 14 and the radiofrequency generator 15 shown in FIG. 33 are not provided. The susceptor electrode 8 and the chamber wall 10 have the same DC potential.
In these plasma processing units, power with a frequency of approximately 13.56 MHz is generally supplied in order to generate a plasma between the electrodes 4 and 8. A plasma process such as a plasma-enhanced CVD process, a sputtering process, a dry etching process, or an ashing process is then performed using the plasma.
The operation validation and performance evaluation of the above-described plasma processing units have been conducted by actually performing the process such as deposition and then evaluating the deposition characteristics thereof according to following Procedures:
Procedure (1) Deposition Rate and Planar Uniformity
Step 1: Depositing a desired layer on a 6-inch substrate by a plasma-enhanced CVD process.
Step 2: Patterning a resist layer.
Step 3: Dry-etching the layer.
Step 4: Removing the resist layer by ashing.
Step 5: Measuring the surface roughness using a contact displacement meter to determine the layer thickness.
Step 6: Calculating the deposition rate from the deposition time and the layer thickness.
Step 7: Measuring the planar uniformity at 16 points on the substrate surface.
Procedure (2) BHF Etching Rate
A resist mask is patterned as in Steps 1 and 2 in (1) above.
Step 3: Immersing the substrate in a buffered hydrofluoric acid (BHF) solution for one minute to etch the layer.
Step 4: Rinsing the substrate with deionized water, drying the substrate, and separating the resist mask using a mixture of sulfuric acid and hydrogen peroxide (H2SO4+H2O2).
Step 5: Measuring the surface roughness as in Step 5 in Procedure (1) to determine the layer thickness after the etching.
Step 6: Calculating the etching rate from the immersion time and the reduced layer thickness.
Procedure (3) Isolation Voltage
Step 1: Depositing a conductive layer on a glass substrate by a sputtering method and patterning the conductive layer to form a lower electrode.
Step 2: Depositing an insulating layer by a plasma-enhanced CVD method.
Step 3: Forming an upper electrode as in Step 1.
Step 4: Forming a contact hole for the lower electrode.
Step 5: Measuring the current-voltage characteristics (I-V characteristics) of the upper and lower electrodes by using probes while applying a voltage up to approximately 200 V.
Step 6: Defining the isolation voltage as the voltage V at 100 pA corresponding 1 xcexcA/cm2 in a 100 xcexcm square electrode.
The plasma processing apparatus has been required to achieve a higher plasma processing rate (the deposition rate or the processing speed), increased productivity, and improved planar uniformity of the plasma processing (uniformity in the distribution of the layer thickness in a planar direction and uniformity in the distribution of the process variation in the planar direction). As the size of substrates has been increasing in recent years, the requirement of planar uniformity has become tighter. Moreover, as the size of the substrate is increased, the power required is also increased to the order of kilowatts, thus increasing the power consumption. Accordingly, as the capacity of the power supply increases, both the cost for developing the power supply and the power consumption during the operation of the apparatus are increased. In this respect, it is desirable to reduce the operation costs.
Furthermore, an increase in power consumption leads to an increase in emission of carbon dioxide which places a burden on the environment. Since the power consumption is increased by the combination of increase in the size of substrates and a low power consumption efficiency, reduction of the carbon dioxide emission is desired.
The density of the plasma generated can be improved by increasing the plasma excitation frequency. For example, a frequency in the VHF band of 30 MHz or more can be used instead of the conventional 13.56 MHz. Thus, one possible way to improve the deposition rate of a deposition apparatus such as a plasma-enhanced CVD apparatus is to employ a higher plasma excitation frequency.
In a plasma processing apparatus having a plurality of the above-described plasma processing units, variation in plasma processing among the plasma processing units and their matching circuits is required to be reduced, so that the plasma processing rate (deposition rate when applied to a deposition process), productivity, and uniformity in the plasma process in the planar direction of a workpiece (planar distribution in the layer thickness) can be made substantially the same among the workpieces plasma-treated in different plasma processing units.
The plasma processing apparatus is also required to yield substantially the same process results by applying the same process recipe specifying external parameters for respective plasma processing units such as gas flow, gas pressure, power supply, and process time.
It is desired to both reduce the time required for adjusting the plasma processing apparatus newly installed or subjected to maintenance to achieve substantially the same process results by applying the same recipe and eliminate the variation among the plurality of plasma processing units, as well as the cost required for such adjustment.
Furthermore, reduction in the variation among the plasma processing units has also been required for a plasma processing system comprising a plurality of such plasma processing apparatuses.
The above-described plasma processing unit is designed to use power with a frequency of approximately 13.56 MHz and is not suited for power of higher frequencies. Specifically, radiofrequency characteristics such as impedance and resonant frequency characteristics of the plasma processing unit as a whole, and more specifically, the radiofrequency characteristics of the plasma processing chamber and the matching circuit have been neglected; consequently, no improvement in the electrical consumption efficiency has been achieved when power of a frequency higher than approximately 13.56 MHz is employed, resulting in decrease in the deposition rate rather than improvement. Moreover, although the density of a generated plasma increases as the frequency increases, the density starts to decrease once its peak value is reached, eventually reaching a level at which glow-discharge is no longer possible, thus rendering further increases in frequency undesirable.
In a plasma processing apparatus and a plasma processing system comprising a plurality of plasma processing apparatuses, the radiofrequency characteristics of the plasma processing units including the matching circuits are defined by their mechanical dimensions such as shape. However, the components constituting the plasma processing unit inevitably have differences in size, etc., due to the mechanical tolerance during manufacture. When these components are assembled to make a plasma processing unit, the tolerance due to the assembly is added to the tolerance in the mechanical dimensions. Furthermore, some portions of the plasma processing chamber may not be measurable after assembly of the components; consequently, whether the plasma chamber as a whole has designed radiofrequency characteristics may not be quantitatively validated. Thus, means for examining the variation in the radiofrequency characteristics of the plasma processing chambers has not been available.
Thus, conventional plasma processing apparatuses suffer from the following disadvantages.
Conventional plasma processing apparatuses and systems are not designed to eliminate the differences in electrical radiofrequency characteristics such as impedance and resonant frequency characteristics among the plasma processing units constituting the plasma processing apparatus or system. Thus, the effective power consumed in the plasma generating spaces of the plasma processing units and the density of the generated plasma vary between different plasma processing units.
As a consequence, uniformity in plasma process results may not be achieved when the same process recipe is applied to these plasma processing units.
In order to obtain uniform plasma process results, external parameters such as gas flow, gas pressure, power supply, process time, and the like must be compared with the process results according to Procedures (1) to (3) described above for each of the plasma processing units so as to determine the correlation between them. However, the amount of data is enormous and it is impossible to completely carry out the comparison.
In order to validate and evaluate the operation of the plasma processing apparatus using Procedures (1) to (3) above, the plasma processing apparatus needs to be operated and deposited substrates need to be examined by an ex-situ inspecting method requiring many steps.
Since such an inspection requires several days to several weeks to yield evaluation results, the characteristics of the plasma-treated substrates processed during that period, supposing that the production line is not stopped, remain unknown during that period. If the performance of the plasma processing apparatus is poor, products not satisfying a required level may be manufactured. In this respect, a method for easily maintaining the operation of the plasma processing apparatus at the required level has been desired.
Moreover, when Procedures (1) to (3) described above are employed to inspect the plasma processing units constituting the plasma processing apparatus or system, the time required for adjusting the plasma processing units so as to eliminate the difference in performance and variation in processing among the plasma processing units to achieve the same process results using the same process recipe may be months. The time required for such adjustment needs to be reduced. Also, the cost of substrates for inspection, the cost of processing the substrates for inspection, the labor cost for workers involved with the adjustment, and so forth are significantly high.
In view of the above, the present invention aims to achieve the following objects.
1. To achieve uniformity in AC resistances as the radiofrequency characteristics in the matching circuits of a plurality of plasma processing units.
2. To achieve uniformity in the plasma process results among the plurality of plasma processing units by applying the same process recipe.
3. To simplify the evaluation of the plurality of the plasma processing units by making it unnecessary to determine process conditions based on the relationships between an enormous amount of data on the plasma processing units and the evaluation results obtained by Procedures (1) to (3) above.
4. To reduce the time required for adjusting the plasma processing units to achieve substantially the same process results using the same process recipe.
5. To improve the planar uniformity of the plasma processing (the planar distribution of the layer thickness or the planar distribution of the process variation) and to improve the layer characteristics of the deposited layers such as isolation voltage when applied to a plasma-enhanced CVD and a sputtering process.
6. To improve the power efficiency and to reduce the power loss so that the same processing rate and layer characteristics are obtained with less power.
7. To reduce the adjustment cost and the operation cost and to increase the productivity.
8. To provide a plasma processing apparatus and system which can be easily maintained at a suitable operation state.
In order to achieve the above-described goals, an aspect of the present invention provides a plasma processing apparatus comprising a plurality of plasma processing units, each of the plurality of plasma processing units comprising: a plasma processing chamber including an electrode for exciting a plasma; a radiofrequency generator for supplying radiofrequency power to the electrode; and a matching circuit for matching the impedances of the plasma processing chamber and the radiofrequency generator, the matching circuit having an input terminal connected to the radiofrequency generator, an output terminal connected to the electrode, and a connection point provided between the input terminal and the output terminal, the matching circuit being connected to a ground potential portion via the connection point. Among the plurality of plasma processing units, a variation  less than RA greater than  defined by equation (14A) below is set at a value within a predetermined range:
 less than RA greater than =(RAmaxxe2x88x92RAmin)/(RAmax+RAmin)xe2x80x83xe2x80x83(14A)
wherein RAmax and RAmin are the maximum and minimum values, respectively, of AC resistances RA in the matching circuits of the plurality of plasma processing units measured from the input-terminal-side of the matching circuits. Also, a variation  less than RB greater than  among the plurality of plasma processing units defined by equation (14B) below is set at a value within a predetermined range:
 less than RB greater than =(RBmaxxe2x88x92RBmin)/(RBmax+RBmin)xe2x80x83xe2x80x83(14B)
wherein RBmax and RBmin are the maximum and minimum values, respectively, of AC resistances RB in the matching circuits of the plurality of plasma processing units measured from the output-terminal-side of the matching circuits.
Thus, the difference in the radiofrequency characteristics, i.e., the AC resistances, affecting the impedance of the matching circuits of the plasma processing units can be minimized and the plasma processing units can be maintained within a predetermined range indicated by the impedance characteristics. Consequently, uniformity in the effective power consumed in the plasma generating spaces of the plasma processing units can be achieved.
When the same process recipe is applied to the plurality of the plasma processing units, substantially the same plasma process results can be obtained. For example, when applied to a deposition apparatus, the deposited layer will exhibit substantially the same layer characteristics such as layer thickness, isolation voltage, and etching rate.
The AC resistances are the radiofrequency characteristics mainly determined by the mechanical structure of the device and are considered to differ among the matching circuits of the plasma processing units. By setting the AC resistances within a predetermined range, the overall radiofrequency characteristics of the plasma processing units which have not been concerned before are optimized, and stable plasma generation can be achieved. Thus, the plasma processing units constituting the plasma processing apparatus or the system incorporating a plurality of plasma processing apparatus stably operate and generate plasmas.
Moreover, examination of the correlation between the external parameters and the inspection results obtained through a conventional inspection method requiring a step of inspecting the actually treated substrates using an enormous amount of data in order to evaluate the process conditions is no longer necessary. Compared to a conventional inspection method requiring the inspection of deposited substrates, the time required for adjusting the plasma processing units to minimize process variations and to constantly achieve the same process results by using the same process recipe can be significantly reduced. The cost of substrates for inspection, the cost of processing the substrates for inspection, and the labor cost for workers involved with the adjustment can also be reduced.
Preferably, the matching circuit is disconnected from the plasma processing unit at the output terminal and at the input terminal, and the AC resistance RA (input-terminal-side AC resistance RA) is measured at a first measuring point corresponding to the input terminal. In this manner, the resistance in the circuit extending from the first measuring point (input terminal) to the connection point on the ground potential portion can be measured.
Thus, the difference in the radiofrequency characteristics of the matching circuits of the plasma processing units can be further minimized, and effective power consumed in the plasma generating spaces of the plasma processing units can be made substantially the same. Compared to the case excluding the matching circuit from the measured region, improved uniformity in the plasma process results using the same process recipe can be achieved.
Preferably, the plasma processing unit further comprises a radiofrequency supplier disposed between the radiofrequency generator and the input terminal of the matching circuit. Preferably, the matching circuit is disconnected from the plasma processing unit at the output terminal and at an input end of the radiofrequency supplier, and the AC resistance RA is measured at a second measuring point corresponding to the input end of the radiofrequency supplier. The difference in the radiofrequency characteristics of both the matching circuits and the radiofrequency suppliers among the plasma processing units can be further minimized compared to the case where the radiofrequency supplier is excluded from the measured region. The uniformity in the effective power consumed in the plasma generating spaces of the plasma processing units can be further enhanced, and highly uniform plasma process results using the same process recipe can be achieved compared to the case in which the radiofrequency supplier is excluded from the measured region.
Preferably, the matching circuit is disconnected from the plasma processing unit at the input terminal and at the output terminal of the matching circuit, and the AC resistance RB is measured at a third measuring point corresponding to the output terminal. The difference in the radiofrequency characteristics of the matching circuits of the plasma processing units can be minimized, and the uniformity in the effective power consumed in the plasma generating spaces of the plasma processing units can be improved. When the same process recipe is applied to these plasma processing units, uniformity in the plasma process results is improved compared to the case in which the matching circuit is not included in the measured region.
Preferably, the plasma processing unit further comprises a radiofrequency feeder disposed between the output terminal of the matching circuit and the electrode. Preferably, the matching circuit is disconnected from the plasma processing unit at the input terminal of the matching circuit and at an output end of the radiofrequency feeder, and the AC resistance RB is measured at a fourth measuring point corresponding to the output end of the radiofrequency feeder. The radiofrequency characteristics in both the radiofrequency feeders (feed plates) and the matching circuits of the plurality of the plasma processing units can be made uniform among these units, and the uniformity in the effective power consumed in the plasma generating spaces of the plasma processing units can be further improved. When the same process recipe is applied to these units, uniformity in the process results is further improved compared to the case excluding the radiofrequency feeder (feed plate) from the measured region.
Preferably, the variations  less than RA greater than  and  less than RB greater than  are less than 0.5 to maintain the uniformity in the plasma process results. When applied to a deposition apparatus, the variation in the layer thickness deposited using the same process recipe among these plasma processing units can be maintained within xc2x17%.
More preferably, the variations  less than RA greater than  and  less than RB greater than  are less than 0.4. The difference in the radiofrequency characteristics, i.e., the impedance, the AC resistance which is the real part of the impedance, the resonant frequency characteristics, the capacitance, etc., of the plasma processing units can be further minimized among these units. Thus, the plasma processing units can be maintained to a predetermined range indicated by the impedance characteristics, and the uniformity in the effective power consumed in the plasma generating spaces of these plasma processing units can be achieved.
As a result, substantially the same plasma process results can be obtained from these plasma processing units when the same process recipe is applied. When applied to a deposition apparatus, the deposited layer will exhibit uniformity in the layer characteristics such as layer thickness, isolation voltage, etching rate, etc. More specifically, when the variations  less than RA greater than  and  less than RB greater than  are less than 0.4, the variation in the thickness of the layers deposited in the different plasma processing units under the same process conditions can be kept within the range of xc2x13%.
Preferably, the AC resistances RA and RB are values measured at a power frequency of the radiofrequency generator. The difference in the radiofrequency characteristics of the plasma processing units during plasma generation can be minimized, and the plasma processing units during plasma generation can be kept within a predetermined range indicated by the impedance. Moreover, the effective power consumed in the plasma generating spaces of these units can be made substantially uniform.
Since resistance R is employed as the indicator, the radiofrequency characteristics at a power frequency can be further directly examined compared to the case in which the impedance Z which is a vector quantity determined by the resistance R and the reactance X is employed.
In this plasma processing apparatus, a radiofrequency characteristic A between the radiofrequency meter and the plasma processing units is set to be the same among these plasma processing units. Thus, the observed values of the AC resistance, impedance, etc., can be deemed equal to the value observed at the measuring point for each of the plasma processing units without correction or conversion.
In order to adjust the radiofrequency characteristics between the radiofrequency meter and each of the plasma processing units to be equal to one another, coaxial cables of the same length connecting the measuring points of the plasma processing units and the radiofrequency meter may be used, for example.
The number of the plasma processing units in a plasma processing apparatus and the number of the plasma processing apparatuses in a plasma processing system can be set as desired.
Moreover, the settings of the radiofrequency characteristics, for example, the AC resistances RA and RB, may differ among the plasma processing units constituting the plasma processing apparatus if the plasma processing units are to perform different plasma processes using different process recipes.
Moreover, the present invention can be applied to a dual frequency excitation plasma-enhanced CVD unit having a first radiofrequency generator, a radiofrequency electrode connected to the first radiofrequency generator, a first matching circuit for matching the impedances of the first radiofrequency generator and the radiofrequency electrode, a radiofrequency-electrode-side matching box for accommodating the first matching circuit, a second radiofrequency generator, a susceptor electrode disposed to oppose the radiofrequency electrode to support a substrate to be treated and connected to the second radiofrequency generator, a second matching circuit for matching impedances of the susceptor electrode and the second radiofrequency generator, and a susceptor-electrode-side matching box for accommodating the second matching circuit. In this unit, the radiofrequency characteristics, such as AC resistances RA and RB, of the second matching circuit can be adjusted in a manner similar to the matching circuit at the plasma excitation electrode side described above.
Preferably, the matching circuit further comprises at least one connection point for connecting the matching circuit to the ground potential portion and the AC resistances RA and RB are measured for each of the connection points by sequentially switching the connection points so that only one of the connection points is connected to the ground potential portion. The variations  less than RA greater than  and  less than RB greater than  among the plurality of the plasma processing units are then defined by equations (14A) and (14B) and adjusted to be within the predetermined range for each of the connection points. Thus, the effective power consumed in these plasma processing units can be made substantially uniform.
Herein, since variations  less than RA greater than  and  less than RB greater than  are calculated for each of the connection points, the largest  less than RA greater than  and  less than RB greater than  are adjusted to be within the predetermined range described above.
Another aspect of the present invention provides a performance validation system for a plasma processing apparatus or system. The performance validation system comprises: a customer terminal; an engineer terminal; and an information provider. The customer terminal requests the information provider via a public line to view performance information indicating the state of operation of the plasma processing apparatus or system described above which a customer purchased from an engineer. The engineer uploads the performance information through the engineer terminal. The information provider provides the performance information uploaded through the engineer terminal to the customer terminal upon the request from the customer terminal. The performance information is provided to a customer considering of purchasing the plasma processing apparatus or system as a basis for making the purchasing decisions. During use of the plasma processing apparatus or system, the performance information is provided to inform the customer of the operation performance and maintenance status.
Preferably, the performance information contains information on the variations  less than RA greater than  and  less than RB greater than  in the AC resistances RA and RB to provide a user with a basis for judging the performance of the plasma processing apparatus or system and to provide a customer considering purchasing the apparatus or system with a basis for making purchasing decisions.
The performance information may be output as a catalog or a specification document.
Another aspect of the preset invention provides an inspection method for a plasma processing apparatus or system described above. The inspection method comprises the steps of: inspecting whether a variation  less than RA greater than  among the matching circuits of the plurality of plasma processing units defined by equation (14A) described above is within a predetermined range; and inspecting whether a variation  less than RB greater than  among the matching circuits of the plurality of plasma processing units defined by equation (14B) above is within a predetermined range. According to this inspection method, whether the uniformity in the radiofrequency characteristics such as impedance, resonant frequency characteristics, and AC resistance is achieved among the plasma processing units can be inspected. Thus, the plasma processing units can be adjusted to be within a predetermined range indicated by the impedance characteristics, and the effective power consumed in the plasma generating spaces of these plasma processing units and the density of the plasma generated in these plasma processing units can be made substantially uniform.
As a result, substantially uniform plasma process results can be achieved in these plasma processing units when the same process recipe is applied. That is, when a deposition process is performed using these plasma processing units, the deposited layers will exhibit substantially the same layer characteristics such as layer thickness, isolation voltage, and etching rate.
The electrical radiofrequency characteristics of each of the plasma processing units and the matching circuits therein are defined by the shape, that is, by the mechanical dimensions. However, the dimensions of each of the components constituting the plasma processing unit vary due to the mechanical tolerance during the manufacturing process. The plasma processing units made by assembling such components inevitably have variations due to both the mechanical tolerance and the assembly tolerance. No method for determining whether the overall plasma chamber has the designed electrical radiofrequency characteristics has been available since some portions are not measurable after assembly of the components. By employing the inspection method of the present invention, the performance of the plasma processing units can be inspected quantitatively and the variation in the radiofrequency characteristics can be examined without measuring the mechanical dimensions. The inspection method is applicable to a plasma processing unit of which the mechanical dimensions are not measurable.
In this inspection method, the input-terminal-side AC resistance RA of the matching circuit may be measured at a first measuring point corresponding to the input terminal of the matching circuit while disconnecting the matching circuit from the plasma processing unit at the output terminal and at the input terminal. Thus, the difference in the radiofrequency characteristics among the plurality of the plasma processing units including the matching circuits can be minimized, the effective power consumed in the plasma generating spaces of these plasma processing units can be made substantially uniform, and higher uniformity in the plasma process results is achieved compared to the case where the radiofrequency characteristics of the matching circuit are not measured.
Instead of the first measuring point described above, a second measuring point may be used to measure the input-terminal-side AC resistance RA of the matching circuit. The second measuring point corresponds to an input end of a radiofrequency supplier disposed between the radiofrequency generator and the input terminal of the matching circuit. The input-terminal-side AC resistance RA is measured at the second measuring point while disconnecting the matching circuit from the plasma processing unit at the output terminal and at the input end of the radiofrequency supplier. Thus, the radiofrequency characteristics of the plasma processing units including the radiofrequency suppliers can be made substantially uniform among these plasma processing units, the effective power consumed in the plasma generating spaces of the plasma processing units can be made substantially uniform, and the higher uniformity in the plasma process results can be achieved compared to the case where the radiofrequency supplier is not included in the measure region.
In this inspection method, the matching circuit may be disconnected from the plasma processing unit at the input terminal and at the output terminal of the matching circuit, and the AC resistance RB may be measured at a third measuring point corresponding to the output terminal. Thus, the radiofrequency characteristics of the plasma processing units can be made substantially uniform among these plasma processing units, the effective power consumed in the plasma generating spaces of the plasma processing units can be made substantially uniform, and higher uniformity in the plasma process results can be achieved compared to the case where the matching circuit is not included in the measure region.
Instead of the third measuring point described above, a fourth measuring point may be used to measure the output-terminal-side AC resistance RB of the matching circuit. The fourth measuring point corresponds to an output end of a radiofrequency supplier disposed between the output terminal of the matching circuit and the electrode. The output-terminal-side AC resistance RB is measured while disconnecting the matching circuit from the plasma processing unit at the input terminal of the matching circuit and at the output end of the radiofrequency feeder. Thus, the difference in the radiofrequency characteristics among these plasma processing units including the radiofrequency feeders (feed plates) can be minimized, the effective power consumed in the plasma generating spaces of these plasma processing units can be made substantially uniform, and higher uniformity in the plasma process results can be achieved using the same process recipe compared to the case where the radiofrequency feeder is not included in the measured region.
In this inspection method, both the predetermined ranges are preferably less than 0.5. In this manner, the uniformity in the plasma process, i.e., whether the variation among these plasma processing units in the thickness of the layers deposited under the same process conditions is maintained within the range of xc2x17%, can be inspected.
In this inspection method, the AC resistances RA and RB are values measured at a power frequency of the radiofrequency generator. Thus, inspection of whether the variation in the radiofrequency characteristics of the plasma processing units in the plasma generating state is eliminated can be conducted. The inspection of whether the plasma processing units in the plasma generating state are maintained in a predetermine range indicated by impedance characteristics, for example, can also be conducted. As a result, the effective power consumed in the plasma generating spaces of the plasma processing units can be made substantially uniform among these plasma processing units.
Since resistance R is employed as the indicator of the radiofrequency characteristics, the radiofrequency characteristics at a power frequency can be further directly examined compared to the case in which the impedance Z which is a vector quantity determined by the resistance R and the reactance X is employed.
In this inspection method, a radiofrequency characteristic between the radiofrequency meter and the plasma processing units is set to be the same among these plasma processing units. Thus, the observed values of the AC resistance, impedance, etc., can be deemed equal to the value observed at the measuring point for each of the plasma processing units without correction or conversion. Thus, the inspection can be performed further efficiently.